Does Leica really need a full frame sensor?

[Olsen]

Unfortunately, Leica is taking a sword to a gun fight. They start at a disadvantage because of the very short lens to film plane distance of the RF. Where Nikon and Canon already have a history with FF sensors, and Nikon now has one in its latest camera, Leica still has to figure out how to make a FF work in the RF. In a year, Nikon and Canon will have even more innovative FF sensors. Leica will always be trailing because they don't have the R&D to compete.

Having said all that, I'm not sure Leica needs to compete. People who buy the M8 are not involved in the dslr wars. They have chosen to follow a different drummer. That's not to say such a consumer is what Leica needs to survive. It is only to reflect where Leica and the M8 now stand. Leica isn't going to take market share from Nikon and Canon regardless of what digital rangefinder they make. Consumers voted with their dollars decades ago about their preference for slrs.

This is very well put.

Please note that both Nikon and Leica is in the same boat. Neither have the resources to develop and make their own sensor. They buy them. Note also how both Canon and Nikon launches newer variants of their fast & wide lenses (now I am repeating myself...). Not without reason. Because the old ones also sends light rays with too high an angle towards the sensor. On DSLRs! So, it would be easier/cheaper/faster for Leica to do the same: Make wide & fast lenses that works better on M8. - After all, Leica is among the best in the camera business when it comes to making optics....

 

[Sofa King]

Leica doesn't necessarily need to compete and has shown in the past that it does not feel the urgency to keep up with the market. They didn't have an in-camera meter until the 80s. They still don't have automatic film advance. They didn't make an electronic shutter until the 90s.

I'd like to see a full frame M. In fact, I've decided not to buy any new digital equipment (I have a D50) unless it's full frame. However, Leica will continue to survive just like they always do with their outdated technology balanced with superior optics, craftsmanship, mechanics, and just the fact that they offer a different style of photographing from the DSLR giants.

 

[JohnL]

On RF cameras, seeking to tap the existing base of M-mount lenses, the FF sensor is vital. Crop sensors are really only good for two things: lower cost and longer effective reach for extreme telephoto lenses (and then only if they match the FF sensor on pixel count, which raises the high-ISO IQ issue) - and RF cameras do not match well with long lenses. On the high-ISO question, a 1.35 sensor cannot be made to match a FF sensor at the "current" technology level, because that would represent an advance of technology which could be applied also to FF sensors, thus bumping them ahead again.
Edit: An afterthought - with a FF format you also have more scope for limiting DOF than with a smaller sensor.

The angle-of-incidence vignetting issue (with RF wideangle lenses) seems potentially close to a solution (in practise), via improved high-ISO capabilities of modern FF sensors with advanced microlens technology coupled with de-vignetting in firmware.

I strongly suspect that the recently-announced Leica upgrade program is partially (at least) a disguised route to eventually incorporating these future developments, without having an eventual M9 make M8s pretty well obsolete as most digital cameras effectively are after very few years.

 

[DWeston]

don't know..

don't know..

considering I have lenses from 15mm to 90mm, equal to 20 to 120mm in a small package, I really don't care what focal lengthts are really, have lenses that have a specific field of view, would rather have frame lines that are accurate for the view actually made by lenses. that and better image quality are my two big wishes....also maybe an easy to use exposure adjustment for when I use A mode. that is it, nothing that keeps me from using it as is....YMMV....

 

[peterm1]

Interesting that I recall only a couple of years ago a representative of one of the major camera companies (I cannot recall which) declaring "that this company will never introduce a full frame sensor." Or equally emphatic words to that effect. I suppose that they have to use such language to inveigle people into buying their current offering - this is all that really matters when you are a camera manufacturer after all.

But now that industry is headed in that direction the train is unstoppable. Of course Leica does not NEED a FF sensor and probably neither do we, but its absolutely inevitable that if this is the technology and marketing trend then this is the direction they have to go too or run the risk of losing out. After all integration of a larger sensor into a camera body does not seem such a huge step once the sensor manufacturers have started making them. Its the making of them that is the difficult bit for obvious reasons.

EDIT and AFTERTHOUGHT : I am sad to say I am old enough to recall the first PCs coming out in the 1980s. The first IBM PC I bought for our office had a 10 - count them - 10 megabyte (Not gigabyte) hard drive and a whole 128 kilobytes of RAM. And that was pretty much state of the art. Many of you will of course recall the progression as RAM, CPU processing power and hard drive storage capacity pretty well doubled every year or two thereafter. The development of cameras is likely to be the same. So far as I can see, its all technology driven or if you prefer, technology in search of a market. Manufacturers in search of ever new markets come up with more powerful and faster models as they develop the capacity to make them. That makes us gear freaks drool uncontrollably, and our eyes glaze over and we rush out to buy upgrades or new kit. For example, the first hard drives seemed in retrospect to be the size of a small two door sedan and could hold not much at all. Now we have solid state devices using flash memory to store gigabytes on a thumb drive. I have every expectation that cameras will do something analogous whether we "need" it or not. We will ultimately find a use for it. On the whole, I suppose its been good that computers have experienced such development and growth as its spawned whole new industries in gaming and video etc as well as producing external benefits in other industries - more efficient cars, digital TV, mobole phones and so on. I suppose it remains to be seen if the same can be said for camera development. Eventually I guess the development of new and better sensors will plateau and manufacturers will focus on some other technology development - maybe even more in-camera "grunt" - ie processing power or some such. Who knows?

 

[RIVI1969]

I agree with peterm1, and by the time Leica release its FF camera, Canon will have FF Rebels for less than 1000 dlls.

 

[DWeston]

I agree with peterm1, and by the time Leica release its FF camera, Canon will have FF Rebels for less than 1000 dlls.

for that matter, how close is the XSi to the image quality of my M8....I bet closer then many would admit, and the XSi has far more functionality....BUT using a RF is a frame of mind for the most part, and my M8 will continue to be useful most likely much longer then I will...or the XSi for that matter. then again, who knows what surprises Leica or Canon, or Nikon or ?? that will further IQ issues alone...

 

[Olsen]

Interesting that I recall only a couple of years ago a representative of one of the major camera companies (I cannot recall which) declaring "that this company will never introduce a full frame sensor." Or equally emphatic words to that effect. I suppose that they have to use such language to inveigle people into buying their current offering - this is all that really matters when you are a camera manufacturer after all.

But now that industry is headed in that direction the train is unstoppable. Of course Leica does not NEED a FF sensor and probably neither do we, but its absolutely inevitable that if this is the technology and marketing trend then this is the direction they have to go too or run the risk of losing out. After all integration of a larger sensor into a camera body does not seem such a huge step once the sensor manufacturers have started making them. Its the making of them that is the difficult bit for obvious reasons.

EDIT and AFTERTHOUGHT : I am sad to say I am old enough to recall the first PCs coming out in the 1980s. The first IBM PC I bought for our office had a 10 - count them, 10 megabyte (Not gigabyte) hard drive and a whole 128 kilobytes of RAM. And that was pretty much state of the art. You will of course recall the progression as both RAM, CPU processing power and hard drive storage capacity pretty well doubled every year or two. The development of cameras is likely to be the same. Its all technology driven. Manufacturers in search of ever new markets come up with more powerful and faster models as they develop the capacity to make them. The first hard drives seemed in retrospect to be the size of a small two door sedan and could hold not much at all. Now we have solid state devices using flash memory to store gigabytes on a thumb drive. I have every expectation that cameras will do something analogous whether we "need" it or not.

But your car still has four wheels...?

There is no real life parell to 'memory' and 'silicium sensors', cost/capacity wise. The price of the sensor quadruples with the size. And there is no hope in the near future that a large sensor is going to be dirt cheap. Like memory.

That it took Nikon 4 years to follow up in Canon's intro of FF-DSLR's is due to that they have no in-house sensor production. Nikon, like Leica, is dependant on that other suppliers can offer this. Silicium sensors has a limitation on the angle of light hitting the sensor. That does not mean that it can't be solved, somehow, but it is expensive and groundbreaking research. Some experts say is impossible. - I am sure that folks working in the sensor business (Kodak, Dalsa, Foveon etc.) frequent photo sites like RFF and know far more than I about this and can contribute.

Full frame sensors have been around - for DSLR's since 2002. If it was easy to put into a rangefinder camera, both Cosina and Leica would have done it already. Be aware: Full frame sensors do not give the same result as on film. Vignetting and unsharp corners is predominant if it is not a 'modern' lense corrected for digital use. You see both Canon and Nikon releasing new versions of their 'wide & fast' lenses so they will perform better on Full Frame DSLRs. Would they do that if a new sensor with a 'vignetting/soft corners cure' was just around the corner..?

 

[ernesto]

The two matters are independent:

1) a crop factor means that the lens are designed for a certain format and the sensor does not match it. When a lens is designed it is done balancing the pros and cons. If you design a lens for a full frame sensor, and you use it with a smaller, you are not being efficient, because you have to accept the limitations of another format that you are not using, and you cannot have the advantages that could result from a smaller sensor. That is why many manufacturers have provided special lens for small sensors.

2) the other matter, is totally independent and ypu can have a full frame sensor and a small with bad noise, and the oposite!

Ernesto

 

[shimo-kitasnap]

does leica need a full frame? I'll only buy a "new" one if it's full-frame. I want my wides and normals to be well, wides and normals. my reason? I don't want a 28mm f2.8 substitue for a 35mm view which is a huge lens compared to the 35/2 or even my 35/25 pancake II.

 

[peterm1]

There is no real life parell to 'memory' and 'silicium sensors', cost/capacity wise. The price of the sensor quadruples with the size. And there is no hope in the near future that a large sensor is going to be dirt cheap. Like memory.

While I cannot claim to be an expert on the technology I am not sure I agree, with all due respect.

With such technologies (as silicon chips, CPUs and yes camera sensors) the cost is in two main areas. (1) research and development - this is where tech companies pour billions into a big black hole to develop new products and also to develop the new technologies needed to reliably mass produce these new products. This is why more development of sensors will possibly be with specialist makers - not the leicas and the nikons of the world. (2) Production facilities. A high proportion of the costs of commercialising such technology are fixed costs. ie in the machines needed to produce the end product. The actual raw material is a smaller part of the equation. The cost is in those huge manufacturing plants with clean rooms and multi million dollar gegaws and doo dads.

In any industry where there are high fixed cost the solution is always the same - mass production to amortise those high fixed costs over a larger production and sales base. The same goes for amortising R and D costs.

So the incentive for chip / sensor people and certainly for the Nikons and the Canons of the world is to get to the mass marketing phase as quickly as possible to make your bundle before technology moves on again. The way this normally happens is that new technologies are introduced into products at the top end, where people are willing to pay for it. Then the technology is dispersed more widely through the consumer mass market and prices begin to fall for the same technology. Eventually yesterdays new big feature is today's baseline technology that everyone takes for granted.

Why would the same not happen here? Sure there will be technical hurdles along the way but history suggests that these can be overcome. I guess they are already being overcome if camera manufacturers are already selling models with FF sensors.

I am not saying BTW that Leica should plan to get into the mass market, that is not its business model and it is not what Leica's name is built on. Rather, I am saying that it is inevitable that other companies' middle and low end cameras will eventually begin to incorporate FF sensors, and before this happens Leica better do so too otherwise its claims to high end quality will begin to look pretty damn silly by oomparison.

 

[shimo-kitasnap]

that makes a lot of sense, I remember when I bought my first flat panel monitor, (it was a 15inch cost about $1k) now they're way way cheaper and better. Canon and Nikon both sold their Eos 10D and D100s at much higher prices than they sell they're D300 and Eos 40D which vastly improve upon the initial designs. Why Leica doesn't follow? Why didn't they follow the SLR trend at the end of the 60s? beats me......why doesn't Ferrari release cheaper more economical cars?

 

[dcsang]

To answer the question posed in the subject of this post:
No.. Leica does not need a full frame sensor; at least not until Cosina/Voigtlander pairs up with someone to create a full frame sensor digital rangefinder. 😀

Remember who made the first digital rangefinder which, obviously, prompted Leica to get it's a$$ in gear. 😀

Dave

 

[msbel]

Leica will introduce new lenses with a FF sensor that are maximized to digital capture in design and weight.

To bank long term on premium $3k+ lenses for customers who want more than the Summarit line-up 'aint going to fly long term.

There is a space between the current Summarit and 28 2.8 "less expensive but still are" lenses and the suite of Lux's reaching the $4K range, and Cron ASPH's not far behind.

My own totally unqualified guess is a new suite of lenses under a M mount "D" moniker that will not require coding, IR filters, etc. will be introduced with a FF by the end of the year. Price point $1800-2200.

Many criticize Leica's marketing prowess, but when I step back and consider I bought a $5000 camera and new lenses for the crop factor, and now would happily pay for a FF sensor upgrade when it comes, and dang, brand new Leica lenses for digital....


Well...not so bad as far as most premium brands.

 

[Artorius]

Am I missing something here?

Am I missing something here?

Leica currently has/had made fast wides for 35mm, CV does too. Why would there be a need for new glass for FF that can already be used for FF 35mm. Isn't it just the distance from the lens to film plane? Put the sensor where the film should be and I would think problem solved, though not an engineer.
Do I need full frame, no. Would I like full frame, maybe. What I would/could use is a sensor with more density, higher ISO less noise, lower ISO and less noise, and a better dynamic range. I would stay with the 1.33 CF if they could do that. I don't know at this time that FF sensors would give me what I want. I've looked at Canon and Nikon(my preference camera) and still don't see anything other than MPs.

Forgot to add, with the new FW, my M8 is still producing what I currently need. The comments above are what I would like in the near future.

 

[Sofa King]

^ The problem with a full-frame sensor as they are currently built is that the pixels are placed at the bottom of tiny wells on the surface of the sensor. Therfore, since the rear element of the lens is so close to the sensor on an RF camera, the light can't reach the bottom of those wells. It's like if you're in a valley during sunrise and the mountains are shading the sun early in the morning because it's at such an angle.

Film doesn't have this problem because it is a flat surface. No wells.

 

[JohnL]

The two matters are independent:

<snip>

2) the other matter, is totally independent and ypu can have a full frame sensor and a small with bad noise, and the oposite!

At any given level of technology, the smaller the pixel, the harder it becomes to get low-noise performance at higher ISO levels.

 

[Ororaro]

Oh no! Not again!

"The Industry", "Bad because the DOF changes" (yeah, like if our eyes and brain could really visualize a photo in FF format but the crop ruins it all), "The Canon Rebel BlaBla".

What a boring subject.

 

[nitrox1]

But as far as the sensor itself goes, I don't have any problem with it, I've gotten used to the crop and the thing holding my images back sure isn't the sensor or any crop factor, it's me.

Me too. That and my digital processing "skills". 🙂

I certainly don't mind the crop factor, didn't mind it on my old dslr. An improved lower noise at high iso sensor would be more important to me. But as it is, I'm very happy with the M8 as it is right now. Of course, I'm still on my honeymoon.


John

 

[Ara Ghajanian]

I couldn't wait for one any longer. I want to shoot, not wish.
Ara

 

[Al Patterson]

Moore's Law agrees with you.

Moore's Law agrees with you.

There is no real life parell to 'memory' and 'silicium sensors', cost/capacity wise. The price of the sensor quadruples with the size. And there is no hope in the near future that a large sensor is going to be dirt cheap. Like memory.

While I cannot claim to be an expert on the technology I am not sure I agree, with all due respect.

With such technologies (as silicon chips, CPUs and yes camera sensors) the cost is in two main areas. (1) research and development - this is where tech companies pour billions into a big black hole to develop new products and also to develop the new technologies needed to reliably mass produce these new products. This is why more development of sensors will possibly be with specialist makers - not the leicas and the nikons of the world. (2) Production facilities. A high proportion of the costs of commercialising such technology are fixed costs. ie in the machines needed to produce the end product. The actual raw material is a smaller part of the equation. The cost is in those huge manufacturing plants with clean rooms and multi million dollar gegaws and doo dads.

In any industry where there are high fixed cost the solution is always the same - mass production to amortise those high fixed costs over a larger production and sales base. The same goes for amortising R and D costs.

So the incentive for chip / sensor people and certainly for the Nikons and the Canons of the world is to get to the mass marketing phase as quickly as possible to make your bundle before technology moves on again. The way this normally happens is that new technologies are introduced into products at the top end, where people are willing to pay for it. Then the technology is dispersed more widely through the consumer mass market and prices begin to fall for the same technology. Eventually yesterdays new big feature is today's baseline technology that everyone takes for granted.

Why would the same not happen here? Sure there will be technical hurdles along the way but history suggests that these can be overcome. I guess they are already being overcome if camera manufacturers are already selling models with FF sensors.

I am not saying BTW that Leica should plan to get into the mass market, that is not its business model and it is not what Leica's name is built on. Rather, I am saying that it is inevitable that other companies' middle and low end cameras will eventually begin to incorporate FF sensors, and before this happens Leica better do so too otherwise its claims to high end quality will begin to look pretty damn silly by oomparison.

I would suggest the PC CPU Chip is a more accurate barometer of where CMOS sensors are going. I'd expect each generation to be twice the number of Pixels at roughly the same cost every 18 months or so. I'm sure we'll see $1,000 FF DSLRs fairly quickly, like maybe by 2010 or so. At that point, maybe someone can make a $1,500 FF DRF.